IOData.two_char_chain — whether to read PDB files with two-character chain IDs

Models containing chain names longer than a single character cannot be represented in traditional PDB format as only a single column is allocated for the chain ID. If this variable is set True, the default, MODELLER will support two-character chain IDs, by using the otherwise-unused column 21 (between the residue name and chain ID). (See also IOData.hybrid36 for similar handling of large residue and atom numbers.) This can be set False to strictly conform to the traditional PDB format, for example when dealing with software such as AMBER that uses column 21 for another purpose, such as longer residue names.

Note that MODELLER will always use this scheme to write out PDB files, as it is the same as traditional PDB for single-character chain IDs. (Thus, to maximize compatibility, avoid two-character chain names.)

Use of the mmCIF format is encouraged when working with large systems.